DA. VIS STRAIT AND LABRADOR SEA 139 



to the continental slope (fig. 113), or it may project southwesterly 

 past the Tail (fig. 115), or run nearly south (fig. 117). Another 

 characteristic type of the circulation west of the Tail which persisted 

 for over 2 months the spring of 1926 is featured by a cyclonic 

 eddy (figs. 105, 106, 107, and 109). The system west of the Tail, 

 on the other hand, may be completely altered in appearance by the 

 encroachment of the Atlantic Current toward the southwest slope 

 (figs. 102, 116, and 117). 



(b) The Labrador Current sometimes meets the Atlantic Current 

 in such a manner that a cold-water salient extends southeasterly 

 from the Tail along the 4,000-meter isobath as far as latitude 41°, 

 longitude 47°. This course of the currents appears to favor the 

 formation of a cyclonic eddy which may travel about southeast of 

 the Tail as shown on figures 106, 107, and 115. An apparent weaken- 

 ing of the Labrador Current may be the cause of the retreat of this 

 eddy to the northward along the eastern slope of the Grand Banks 

 (figs. 105, 112, 117, 118, and 120). 



((?) The previously described intrusion of the Atlantic Current 

 between the forty-fourth and forty-fifth ]3arallels on the east side 

 of the Grand Banks is one of the most constant features of the 

 circulation (figs. 102, 109, 110, 112, 116, 117, 120, and 121). This 

 development of the Atlantic Current during the latter part of the 

 iceberg season and the consequent deflection of the ice from the main 

 steamship tracks has occasionally been a determining factor in the 

 discontinuation of the Ice Patrol for that season. 



(d) The Labrador Current upon crossing the Flemish Cap Ridge 

 often gives off a small branch southeastward between the northern 

 border of the Atlantic Current and Flemish Cap (figs. 112 and 116). 



(e) The system of circulation at the junction of the Labrador 

 Current and the Atlantic Current sometimes becomes very complex 

 as attenuated sinuous tongues of superficial current interlace. The 

 cold southwesterly current, for example, which intersected the plane 

 of the Michael Sars^ section south of the Grand Banks in June 1910 

 (Helland-Hansen, 1930). indicated a system of circulation similar to 

 that shown on figures 109 and 110. The surface temperature maps 

 compiled fortnightly by the Ice Patrol during the ice season, each 

 based upon hundreds of temperature reports from passing steamers, 

 indicate that narrow tongues and occluded pools of warm and cold 

 surface water from the two main currents sometimes extend con- 

 siderable distances on both sides of the boundary as marked by the 

 dynamic topographic maps. The more closely the subsurface obser- 

 vations are taken in the active mixing area, it is safe to state, the 

 more complex the currents will be revealed there. 



The variation in the volume of the Labrador Current and the 

 Atlantic Current in the Grand Banks sector (1910-35) is shown by 

 the table on page 143. The computed volumes of the Labrador Cur- 

 rent during the earlier part of the period are not so accurate as the 

 values shown for the later years when the station's observations were 

 taken at much closer intervals. The series unfortunately is not of suf- 

 ficient length or continuity to determine whether or not a correlation 

 exists between the volume of the Labrador Current and the num- 

 ber of icebergs south of Newfoundland. If there be any direct rela- 

 tion, it apparently is masked by frequent and irregular fluctuations, 



79920—37 10 



